A New Neurocognitive Model for Assessing Divergent Thinking: Applicability, Evidence of Reliability, and Implications for Educational Theory and Practice

Abstract

A new model is presented to assess divergent thinking from a neurocognitive perspective. Evidence is presented showing its sensitivity to capture differences in divergent thinking. This observational interview-based study used a two-dimensional matrix (grid) composed of dimensions derived from neuropsychology and cognitive science theory. Each category (grid cell) is related to a particular cognitive operation that putatively is mediated by one or more functional/anatomical brain regions. Using the coding matrix, data were obtained from 30 middle school students through interviews and think-aloud responses to questions. Students were videotaped while viewing rich pictorial material and responses were coded and characterized for divergent production using 10 coefficients: higher category responses, detail, spontaneous generation, perspective switching, category switching per probe, category switches per utterance, different switches, detail per switching ratio, new subjects, higher category responses and reasoning. Reliability was obtained by interobserver concordance and Cohen's kappa coefficients were calculated for the cognitive and brain functional references dimensions (.90 and .91, respectively). Individual differences were expressed as variations in divergent thinking, providing a convenient way of assessing divergent thinking that may be related to creative thinking in education. This model may have heuristic value beyond research on creativity by providing a theoretical framework to guide further research on the cognitive correlates of divergent thinking with brain functional modules by correlating brain imaging data with concurrent interview data gathered by use of the model.

We thank Dr. Peter Gordon, Dr. Felicia Moore, Dr. James Corter, and Dr. Kim Kastens of Columbia University and Teachers College for expert advice during our research and preparation of this article. This material is based upon work supported in part by the National Science Foundation under Grant No. 0350288 (O. R. A., P.I.) and a Spencer Foundation Grant to C. B. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Notes

Note. D = detail provided for each probe; C = category switching, moving from one cell to another, across the matrix; H = higher cognitive category responses; P = perspective switching, percent of total responses involving a switch in perspective; NS = the proportion of times that a new subject was introduced in the narrative; S = amount of spontaneously generated information; CU = category switches per utterance; DS = number of different brain area categories involved in switching; R = extent to which individuals used reasoning.